J. Chem. Eng. Data 2009, 54, 3079–3081
3079
Adsorption of Nitrous Oxide on Activated Carbons Yong Peng, Fumin Zhang, Chunhui Xu, Qiang Xiao, Yijun Zhong, and Weidong Zhu* Key Laboratory of the Ministry of Education for Advanced Catalysis Materials, Institute of Physical Chemistry, Zhejiang Normal University, Jinhua 321004, P. R. of China
The adsorption equilibria of nitrous oxide (N2O) on three commercial activated carbons were systematically investigated by means of a volumetric method at pressures up to 101 kPa and temperatures ranging from (195 to 323) K. The multiple-process adsorption equilibrium model appropriately describes the equilibrium data over the whole range of the applied conditions. The estimated total adsorption capacity of N2O is higher for activated carbon which possesses a higher pore volume determined by N2 adsorption at 77 K. The thermodynamic properties such as the adsorption equilibrium constant and enthalpy associated with adsorption were derived to characterize interactions between adsorbate and adsorbent.
Table 1. Textural Properties of the Activated Carbons Investigated
Introduction Nitrous oxide (N2O) is a strong greenhouse gas and has aroused great attention.1,2 The potential for global warming and destruction of ozone has spurred the development of various strategies to decrease the concentration of N2O in the atmosphere. One of the methodologies is of direct decomposition over effect catalysts, either by an end-of-pipe treatment or by the reuse of N2O in the tail gas for other reactions to other valuable products.2-4 An adsorption-based process is an alternative to the recovery of N2O in nitric acid plants where diluted N2O streams (0.05 % to 0.2 % by volume) are present to produce a concentrated stream (> 5 % by volume) or to store N2O which is characterized by high concentrations (> 30 % by volume) in exhausts from adipic acid.5 Activated carbon or post-treated activated carbon is a promising adsorbent due to its unique advantages such as large specific surface area and inherently large pore volume, resulting in a large adsorption capacity.6 Therefore, activated carbon is the most widely used adsorbent in adsorption-based processes.6,7 For industry applications of gas adsorption and recovery as well as in the characterization of porous activated carbon, it is essential to acquire information on adsorption equilibria, which is a prerequisite to understand an adsorption process. Experimentally, however, only limited work has been carried out on a comparative and systematic investigation on the adsorption of N2O on different types of activated carbon, although several hundred sets of data for numerous adsorbates on carbon materials have been measured, and these data are accumulated in a handbook by Valenzuela and Myers.8 This paper presents experimental results for the adsorption isotherms of N2O on three commercial activated carbons determined by a volumetric method. In addition, thermodynamic properties such as the adsorption equilibrium constant and enthalpy associated with adsorption are determined to characterize interactions between adsorbate and adsorbent.
Experimental Section Adsorption Measurements. A Micromeritics ASAP 2020 gas adsorption analyzer (stainless steel version) was used to ac* Corresponding author. Tel./Fax: +86-579-82282932. E-mail: weidongzhu@ zjnu.cn.
adsorbent
SBET/m2 · g-1
Vtotal/cm3 · g-1
Vmicro/cm3 · g-1
Kureha Ovcls Vruf
1300 1055 1330
0.560 0.501 0.810
0.560 0.397 0.040
curately measure the adsorption isotherms of N2O on three adsorbents in the pressure range from (0.01 to 101) kPa. The instrument was equipped with a molecular drag vacuum pump and two different pressure transducers to enhance the sensitivity in the low-pressure range, which is especially useful in adsorption studies on adsorbents containing micropores. The static-volumetric technique was used to determine the volume of gas adsorbed at different partial pressures: upon adsorption, a pressure decrease was observed in the gas phase, which was a direct measure of the amount adsorbed. The sample cell was loaded with approximately 200 mg of the adsorbent particles. Prior to the adsorption measurements, the adsorbent particles were outgassed in situ in a vacuum at 623 K for 16 h to remove any adsorbed impurities. The obtained dry sample mass was used in the calculation of isotherm data. Adsorption measurements were subsequently carried out at four different temperatures (195 K, 273 K, 293 K, and 323 K). The temperatures were controlled within ( 0.3 K using solid carbon dioxide and an ice-water mixture as coolants to fix the temperatures at (195 and 273) K, respectively, and a circulating water bath was used to control the temperatures at (293 and 323) K within ( 0.1 K. Adsorbents and Adsorbates. Three commercial activated carbons were used as the adsorbents in this study. They were purchased from the Kureha Chemical Industry and Calgon Carbon Corporation (brand codes: OVC-LS and VRU-F). These activated carbons are referred to as Kureha carbon, Ovcls carbon, and Vruf carbon, respectively. The textural properties of Kureha carbon were reported elsewhere.9 Kureha carbon is a purely microporous material. The textural properties of Ovcls carbon and Vruf carbon were determined by the adsorption isotherms of N2 at 77 K measured with a Micromeritics ASAP 2020 gas adsorption analyzer. The textural properties of all the investigated activated carbons are summarized in Table 1. Ovcls carbon contains primarily micropores and a small quantity of mesopores, while Vruf carbon
10.1021/je9001625 CCC: $40.75 2009 American Chemical Society Published on Web 05/29/2009
3080
Journal of Chemical & Engineering Data, Vol. 54, No. 11, 2009
Figure 1. Adsorption isotherms of N2O on Kureha carbon. Lines are the threeprocess adsorption equilibrium model correlation. [, 195 K; 9, 273 K; 2, 293 K; b, 323 K. The isotherm at 195 K corresponds to the left y-axis, and the isotherms at other temperatures correspond to the right y-axis.
Figure 3. Adsorption isotherms of N2O on Vruf carbon. Lines are the threeprocess adsorption equilibrium model correlation. [, 195 K; 9, 273 K; 2, 293 K; b, 323 K. The isotherm at 195 K corresponds to the left y-axis, and the isotherms at other temperatures correspond to the right y-axis.
Figure 2. Adsorption isotherms of N2O on Ovcls carbon. Lines are the threeprocess adsorption equilibrium model correlation. [, 195 K; 9, 273 K; 2, 293 K; b, 323 K. The isotherm at 195 K corresponds to the left y-axis, and the isotherms at other temperatures correspond to the right y-axis.
Figure 4. Adsorption isotherms of N2O on Kureha carbon correlated by the To´th model. [, 195 K; 9, 273 K; 2, 293 K; b, 323 K. The isotherm at 195 K corresponds to the left y-axis, and the isotherms at other temperatures correspond to the right y-axis.
possesses both micro- and mesopores and only a small amount of micropores contribute to its total pore volume. The gaseous adsorbate nitrous oxide with a purity over 99.99 % was used without further purification. Ultrahigh purity (> 99.999 %) helium was used to determine the free space of the sample cell. Isotherm Model. For practical utility, the experimental data should be described by an analytical expression that includes temperature dependency. Various empirical or semiempirical models have been proposed to fit isotherm data in adsorption science. The multiple-process adsorption equilibrium model, which was introduced by Drago et al.10-13 to analyze gas adsorption equilibria in porous carbonaceous materials and zeolites, is based on the fact that an adsorbent consists of several different size pores, and it is reasonable to premise that adsorption involves a series of elementary equilibrium processes that differ in their enthalpy and entropy. Each process may be a heterogeneous grouping of pores whose affinities are too close to be discriminated by the accuracy of experimental measurements. As each equilibrium process is described by an equilibrium constant, Ki, and a finite capacity, qi, the total adsorption isotherm is given by eq 1
pressure. The quantity qsat is temperature independent as it i represents a saturation capacity for one adsorption process. The equilibrium constant Ki is temperature dependent and effectively describes the extent to which a particular type of pore fills at a given temperature. The parameter Ki taking the van’t Hoff relation for the adsorption affinity can be written as
n
q)
qsatK p
∑ 1 +i Ki ip
(1)
i)1
where q is the total amount adsorbed; qsat i is an intrinsic property of the adsorbent reflecting the capacity for process i; Ki is the associated equilibrium constant; and p is the equilibrium
[
Ki ) Ki0 exp
(
)]
-∆Hads,i T0 -1 RgT0 T
(2)
where Ki0 is the adsorption affinity at reference temperature T0; ∆Hads,i is a measure of the enthalpy of adsorption; Rg is the universal gas constant; and T is the temperature studied. The lowest temperature investigated, 195 K, is used as T0, since the measured isotherm at this temperature covers the widest range of loading.
Results and Discussion The adsorption isotherm data of N2O on the three commercial activated carbons are given by the points in Figures 1 to 3. All the isotherms exhibit a type-I adsorption isotherm (Brunauer classification) over the temperature and pressure range investigated. The measured isotherm data were correlated by the multipleprocess adsorption equilibrium model. The nonlinear parameter estimation was carried out in two steps using a combined fitting method.14 Three processes were required to obtain good fits to the experimental data. First, the six parameters in eq 1, referred to as q1sat, q2sat, q3sat, K10, K20, and K30, were extracted from the isotherm data at T0, i.e., 195 K. Second, the parameter ∆Hads,i for each adsorption process was estimated from the isotherm data at all the
Journal of Chemical & Engineering Data, Vol. 54, No. 11, 2009 3081 Table 2. Estimated Values of Adsorption Isotherm Parameters in the Three-Process Adsorption Equilibrium Model and Their Standard Deviations (95 % Confidence Limit) parameter qsat/mol · kg-1
K0/kPa-1 -∆Hads/kJ · mol-1
process 1 2 3 Total 1 2 3 1 2 3
Kureha carbon 1.109 ( 0.223 4.027 ( 0.502 10.183 ( 0.610 15.319 ( 1.335 27.760 ( 10.065 0.902 ( 0.266 0.092 ( 0.007 28.978 ( 0.237 25.389 ( 0.280 23.095 ( 0.303
temperatures by eqs 1 and 2. The isotherm data can be appropriately described by the three-process adsorption equilibrium model, as shown in Figures 1 to 3 by the drawn lines. Other isotherm models, such as the To´th, Unilan, and Dubinin-Radushkevich (DR) models,15 which are frequently used to describe adsorption equilibria of various adsorbates on different types of activated carbon, have also been considered. The Unilan and DR models deviate from the measured isotherm data to a great extent in the current case. In comparison with the Unilan and DR models, the To´th model gives a relatively better description for N2O adsorption on the three activated carbons investigated. Figure 4 shows an example for the isotherms of N2O on Kureha carbon correlated by the To´th model. However, as seen from Figure 4, the line predicted by the To´th model is slightly below the experimental data points at 195 K in the lower pressure range, while at moderate pressures the correlated isotherm is somehow higher. This could be due to the high degrees of heterogeneity for the activated carbon materials investigated, and the To´th model with three fitting parameters cannot properly describe this heterogeneity, resulting in the deviation of model predictions from isotherm data. In summary, the three-process adsorption equilibrium model describes the present case the best over the full range. The values of qsat i , Ki0, and -∆Hads,i extracted by the threeprocess adsorption equilibrium model are summarized in Table 2. The process capacity, qsat i , reflects the number of moles of adsorbate that can be adsorbed by pores with a specific width or range of widths. Summing the adsorption capacities for all the three processes results in the total saturation amount sat sat , which is also presented in Table 2. The qtotal adsorbed, qtotal value is in the order of adsorbent as follows: Vruf carbon > Kureha carbon > Ovcls carbon, which is in line with the increasing total pore volume of these adsorbents (see Table 1). The enthalpy of adsorption, ∆Hads,i, describes the extent of adsorption affinity. The extracted -∆Hads,i value decreases along with the adsorption process. On the same adsorbent, adsorption first occurs in micropores due to the superimposed interaction potentials from opposite surfaces within the carbon pores that generate the significantly larger value of -∆Hads,1. Along with adsorption, the adsorption affinity decreases gradually and generates the smaller -∆Hads,2 value and the even smaller -∆Hads,3 value. The value of -∆Hads,i for processes 1, 2, and 3, presented in Table 2, gradually decreases for N2O on all three activated carbons investigated. This further supports the consideration of the change of adsorption potential along with the adsorption process. To recover N2O in nitric acid plants where diluted N2O streams (0.05 % to 0.2 % by volume) are present,5 Kureha carbon and Ovcls carbon could be suitable adsorbents because they possess mainly micropores, in which the interactions between N2O and pore walls are stronger, resulting in higher amounts adsorbed at low partial pressures of the adsorbate. On
Ovcls carbon 1.887 ( 0.188 6.107 ( 0.647 4.665 ( 0.692 12.659 ( 1.527 39.813 ( 9.828 0.665 ( 0.142 0.062 ( 0.015 29.708 ( 0.126 26.777 ( 0.542 18.427 ( 1.146
Vruf carbon 0.783 ( 0.084 5.525 ( 0.180 30.622 ( 0.751 36.930 ( 1.014 19.979 ( 5.027 0.284 ( 0.026 0.007 ( 0.000 28.246 ( 0.145 25.671 ( 0.368 16.496 ( 0.356
the other hand, to store N2O that is characterized by high concentrations (> 30 % by volume) in exhausts from adipic acid,5 Vruf carbon, which possesses mainly mesopores, could be an alternative adsorbent because it exhibits the greatest adsorption capacity for N2O.
Conclusions The adsorption isotherms of N2O on Kureha carbon, Ovcls carbon, and Vruf carbon have been accurately measured up to 101 kPa and at temperatures in the range of (195 to 323) K. A three-process adsorption equilibrium model gives the best description of the measured adsorption isotherm data on all three adsorbents investigated. The present results might be useful for modeling adsorption-based processes to concentrate tail-gas streams containing diluted N2O or to recover N2O with high concentrations in exhausts from chemical plants.
Literature Cited (1) Rodhe, H. A comparison of the contribution of various gases to the greenhouse effect. Science 1990, 248, 1217–1219. (2) Pe´rez-Ramı´rez, J. Catalyzed N2O activation: promising (new) catalysts for abatement and utilization; Ph.D. Thesis, Delft University of Technology: The Netherlands, 2002. (3) Pe´rez-Ramı´rez, J.; Kapteijn, F.; Mul, G.; Moulijn, J. A. Superior performance of ex-framework FeZSM-5 in direct N2O decomposition in tail-gases from nitric acid plants. Chem. Commun. 2001, 693–694. (4) Panov, G. I. Advances in oxidation catalysis: oxidation of benzene to phenol by nitrous oxide. CATTECH 2000, 4, 18–32. (5) Groen, J. C.; Pe´rez-Ramı´rez, J.; Zhu, W. Adsorption of nitrous oxide on silicalite-1. J. Chem. Eng. Data 2002, 47, 587–589. (6) Yang, R. T. Adsorbents: fundamentals and applications; John Wiley & Sons: NJ, 2003. (7) Bansal, R. C.; Goyal, M. ActiVated carbon adsorption; Taylor & Francis: Boca Raton, 2005. (8) Valenzuela, D. P.; Myers, A. L. Adsorption equilibrium data handbook; Prentice-Hall: NJ, 1989. (9) Zhu, W.; Groen, J. C.; Kapteijn, F.; Moulijn, J. A. Adsorption of butane isomers and SF6 on Kureha activated carbon. 1. Equilibrium. Langmuir 2004, 20, 5277–5284. (10) Drago, R. S.; Burns, D. S.; Lafrenz, T. J. A new adsorption model for analyzing gas-solid equilibria in porous materials. J. Phys. Chem. 1996, 100, 1718–1724. (11) Drago, R. S.; Kassel, W. S.; Burns, D. S.; McGilvray, J. M.; Showalter, S. K.; Lafrenz, T. J. Multiple equilibria analyses of gas-porous solid isotherms. J. Phys. Chem. B 1997, 101, 7548–7555. (12) Drago, R. S.; Webster, C. E.; McGilvray, J. M. A multiple-process equilibrium analysis of silica gel and HZSM-5. J. Am. Chem. Soc. 1998, 120, 538–547. (13) Webster, C. E.; Cottone, A.; Drago, R. S. Multiple equilibrium analysis description of adsorption on Na-Mordenite and H-Mordenite. J. Am. Chem. Soc. 1999, 121, 12127–12139. (14) Zhu, W.; Groen, J. C.; van Miltenburg, A.; Kapteijn, F.; Moulijn, J. A. Comparison of adsorption behaviour of light alkanes and alkenes on Kureha activated carbon. Carbon 2005, 43, 1416–1423. (15) Do, D. D. Adsorption analysis: equilibria and kinetics; Imperial College Press: London, 1998. Received for review February 8, 2009. Accepted May 16, 2009.
JE9001625
